Category: Kaitlyn’s Status Reports

This past week, I continued work on the teams design report submission. I spent a considerable amount of time working on different parts of the report, formatting, and proofreading in order to ensure that our report met all necessary criteria and can carry over to the final report submission.

I also spent time working on the pressure sensor. I wrote all necessary code in order to connect the 16 sensors to the RPi. However, I then had to pivot in order to deal with the fact that there were only 2 addresses possible to be used with the ADCs instead of 4 as originally planned. In order to counteract this, I was able to find an analog 16-1 MUX, which I then attached to one pin of one ADC. Now, instead of using 16 different pins of the different ADC, I will be using one pin of the ADC, and using gpio outputs of the RPi as the select lines to the MUX, which will allow me to iterate through all 16 sensors. Furthermore, I began looking into the script necessary to send the data to the RPi server. In order to do this, I found that we could either use node or flask. Since Cora is more comfortable with node, and I do not have much experience with either, I have started writing the necessary http requests in order to send the data to the server. Our pivot plan in place is to switch to using flask if that leads to easier integration.

My progress is on schedule, as I used this break in order to catch up and finish up the preliminary work on the pressure sensor. After some more thorough testing to ensure it detects proper weight shift, I will sow up a mat in order to make it more movable / less obtrusive to the user.

This week, I hope to finish up testing of the pressure mat, and also work with Cora and Lilly to integrate sending the data to the server, so everyone can work with their necessary components of the project. I also hope to run some timing tests on the mat in order to ensure that it still meets latency requirements.

This weeks deliverables: 

This week,  I worked on the Design Paper which is due this upcoming Friday. Specifically, I worked on improving my block diagrams and making a much more in-depth Gantt chart that we could add to the report.

I also worked on the RasPi and the seat pressure sensors this week. I was able to re-flash Debian onto the Pi and get it connected to the CMU Device network. After that, I installed and configured nginx so that the pi is now hosting a very basic website: 

I was also able to assign it a static IP so that it can be accessed from other networks, so we can test while not on CMU wifi.

I also focused on the sensors, which arrived on Thursday. Since they arrived later than anticipated, I was able to conduct a preliminary test, just to ensure they were working as anticipated.  This was done just using a multimeter and a basic voltage divider circuit. Now that I know that the sensors work, I can focus on wiring them up to the ADC. I was also able to find better documentation on the adafruit ADC I purchased, which help me write some preliminary code to connect the ADC to the rasPi.

 

Progress:

Due to the late arrival of the parts, I am slightly behind schedule on connecting up my sensors to the rasPi. I plan on working a bit extra on the project this week in order to catch up, and anything I am unable to do this week will also be worked on over Spring Break to ensure that I come back ahead of schedule or at least on time.

Deliverables:

This upcoming week, I am focused on connecting the ADCs, RasPi, and the sensors all together. I want to have the code written which can analyze the pressure of the person and at least return a digital value which can be used for further analysis.

This week, I focused on finalizing design choices, ordering parts, and creating the presentation slides/ document.  I spent a lot of time on the force sensitive resistors, because after our meeting with Tamal on Tuesday I realized it would make more sense to go with sensor arrays than with individual resistors – as this would lead to a lot of complicated wiring. I went with the following sensor array:

It is commonly used for measuring weight distribution when standing, and by using multiple of them, I can get a lot more measurement for the same price. Specifically, 1 individual FSR is  the same price as the 8 FSRs that come on this mat. I will not use all of the sensors (because I need to mux them in order to convert them to digital values), but I can play around with the different configurations of which sensor to use. As seen in the picture I am going to use 4 sensors from each pressure mat. From first glance, I want to use the 4 sensors in boxes, but based on testing I will do once the parts arrive, different mats will use different sensors. The current plan is to hook up 4 sensors through a voltage divider circuit into the ADC, which has an op amp amplifier. It will then get sent through I2C communications to the RasPi. I also spent time making the slides for the presentation this upcoming Monday. Outside of that, since we have our RasPi, I am also working towards putting the RasPi on the CMU Device network in order to host the server – this is a more complicated process than anticipated but I am working through it.

I am not behind schedule. This week, I accomplished ordering all of the necessary parts/finalizing BOM & beginning to start hosting the local server, which was what I wanted to accomplish. This upcoming week depending on when my parts arrive I might find difficulty finishing the work I want to do (a first test of the sensors and connecting the ADC) if the parts are delayed / the RasPi does not boot. To keep myself on track, I plan on continuing testing with a RasPi Pico in order to still be able to start writing and testing code for ADC and I2C this week, as this way I can stay on top of the sensor creation. It will also help me start working on the integration earlier.

This week, I presented our Project Proposal to the other capstone teams and faculty. I spent ~2 hours preparing for the presentation to ensure we met all the needed requirements in the slide show, and to ensure that I could spend my time interacting with the audience instead of reading just off the slides. After presenting the proposal, I spent the rest of the week doing research into different sensors we could use in the chair. I was able to narrow it down to two types of sensors: Load Cells and Flexible Force Sensors. After putting some more thought into the overall integration, I think that using flexible force sensors (image of circuit I would need below) might be better in the long run to meet the user requirements of not impeding the user during their work, as they will not be as obtrusive on the chair. While I would need an opamp and capacitors for this circuit, I think the wiring could be done under the chair to be unobtrusive. As well, those parts are very cheap and easy to replace. Since we are using the RasPi, I also researched possible ADC converter modules for the RasPi in order to digitize the voltage inputs into the Pi. I hope this upcoming Monday to finish talking through this design choice with the team and order the parts necessary to begin creating a barebones pressure mat and testing it.

 

My progress is on schedule. This week, I wanted to complete research into parts for the pressure sensors so they could be ordered on Monday when our team next meets.

In the next week, I hope to create a much more in depth design of the chairs sensors. I hope to create a document that details exactly how the pressure sensors will be connected to the raspberry pi, and begin working on the code in order to create the averages for weight distribution on the chair.